INDEX

Explanations

an unprecedented rate

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

1.38

1.27

1.25

Jos

1.23

 alguma

1.19

1.15

 এসেছিল

1.15

Io

1.14

Later

1.13

POSITIVE LOGITS

 食品

1.30

 firearms

1.28

 weapons

1.27

簾

1.26

 dictated

1.25

 criminality

1.24

헤

1.24

 retention

1.23

 dictate

1.22

 printf

1.22

Activations Density 0.002%